Glioblastomas (GBMs) remain among the most devastating of all known human tumors, with median survival times remaining around 12-15 months from initial diagnosis. The introduction of temozolomide chemotherapy, when used concurrently and adjuvantly with radiation, has been shown to significantly improve median survival times and increase the percentage of longer-term GBM survivors. The Radiation Therapy Oncology Group (RTOG), which is one of the largest and most established cooperative groups in oncology, has developed a recursive partitioning analysis (RPA) model for malignant glioma patients, using primarily clinical and demographic variables to stratify malignant glioma patients into one of six distinct prognostic classification groups. The RTOG RPA has long been considered the international "gold standard" as a prognostic model for GBM patients. However, since the original RTOG RPA was developed in the 1990's, two major developments have transpired. First, there has been a rapid advancement in the understanding of the molecular, genetic, and epigenetic mechanisms underlying the pathophysiology and the observed treatment resistance of GBMs, with single institution and limited cooperative group data suggesting that a subset of these biomarkers could serve as useful prognostic markers in GBM. Second, there has been a shift in the adjuvant treatment paradigm of GBMs away from radiation alone (when the original RTOG RPA model was developed) to radiation combined with concurrent and adjuvant temozolomide. Therefore, given these two paradigm shifts, it becomes essential to refine and/or redevelop an RTOG RPA model that is updated along these lines. It is our hypothesis that inclusion of these promising biomarkers will serve to significantly refine the existing RTOG RPA classification model to establish distinct prognostic groups of GBM patients treated in the TMZ era, based on a combination of molecular and clinical variables. The revised RTOG RPA resulting from the proposed effort may be universally used to determine patient eligibility in future clinical trials, as well as to provide guidance with regards to future directions for molecularly-based targeted therapies for GBM patients. The revised RTOG RPA model can be used to establish the "gold standard" expected outcomes for various the prognostic groups of GBM against which the results from clinical trials involving investigational therapies can be compared, much like its decade's old RTOG RPA predecessor model. Therefore, this proposed endeavor is of the utmost importance and relevance for this patient population and will be universally utilized in the international brain tumor community. With regards to methodology, this proposal represents a joint effort between the RTOG and the laboratories of Arnab Chakravarti, MD, Chair and Professor of Radiation Oncology at the Arthur G. James Comprehensive Cancer Center of the Ohio State University and Kenneth Aldape, MD, Professor of Neuropathology at the MD Anderson Cancer Center. Drs. Chakravarti and Aldape are Chair and Co-Chair of the RTOG Brain Tumor Translational Research Group, respectively. Our strategy will be to utilize biorepository specimens from the recently completed RTOG 0525 to accomplish our stated objectives. RTOG 0525 was a Phase III Study conducted in North America, Europe, and Asia comparing dose-dense versus standard dose TMZ when combined with radiation for newly-diagnosed GBM. Tissue blocks were prospectively collected on each and every one of the 1173 GBM patients enrolled on this RTOG study, which have been made available for our NIH challenge grant effort. We will revise the existing RTOG RPA classification model to include not only clinical/demographic variables, but also key molecular, genetic, and epigenetic variables. To this end, we shall validate key signal transduction biomarkers, genetic, and epigenetic markers that have been previously shown to be of prognostic value in smaller GBM studies by our group and others. This data will be combined with the clinical and demographic data previously found to be of importance in the previous RTOG RPA model to generate a revised RPA classification model pertinent to TMZ-treated GBM patients and one that is refined to include molecular, genetic, and epigenetic data of significance.